Dataverse-Python-Pandas-Integration

# Install core dependencies
pip install PowerPlatform-Dataverse-Client
pip install azure-identity

# Install pandas for data manipulation
pip install pandas

from azure.identity import InteractiveBrowserCredential
from PowerPlatform.Dataverse.client import DataverseClient
import pandas as pd

# Setup authentication
base_url = "https://<myorg>.crm.dynamics.com"
credential = InteractiveBrowserCredential()
client = DataverseClient(base_url=base_url, credential=credential)

# Query data
pages = client.get(
    "account",
    select=["accountid", "name", "creditlimit", "telephone1"],
    filter="statecode eq 0",
    orderby=["name"]
)

# Collect all pages into one DataFrame
all_records = []
for page in pages:
    all_records.extend(page)

# Convert to DataFrame
df = pd.DataFrame(all_records)

# Display first few rows
print(df.head())
print(f"Total records: {len(df)}")

# All query parameters return as columns in DataFrame
df = pd.DataFrame(
    client.get(
        "account",
        select=["accountid", "name", "creditlimit", "telephone1", "createdon"],
        filter="creditlimit > 50000",
        orderby=["creditlimit desc"]
    )
)

# Result is a DataFrame with columns:
# accountid | name | creditlimit | telephone1 | createdon

import pandas as pd
from azure.identity import InteractiveBrowserCredential
from PowerPlatform.Dataverse.client import DataverseClient

client = DataverseClient("https://<myorg>.crm.dynamics.com", InteractiveBrowserCredential())

# Load account data
records = []
for page in client.get("account", select=["accountid", "name", "creditlimit", "industrycode"]):
    records.extend(page)

df = pd.DataFrame(records)

# Explore the data
print(df.shape)           # (1000, 4)
print(df.dtypes)          # Data types
print(df.describe())      # Statistical summary
print(df.info())          # Column info and null counts
print(df.head(10))        # First 10 rows

# Filter rows by condition
high_value = df[df['creditlimit'] > 100000]

# Select specific columns
names_limits = df[['name', 'creditlimit']]

# Multiple conditions
filtered = df[(df['creditlimit'] > 50000) & (df['industrycode'] == 1)]

# Value counts
print(df['industrycode'].value_counts())

# Sort by column
sorted_df = df.sort_values('creditlimit', ascending=False)

# Group by and aggregate
by_industry = df.groupby('industrycode').agg({
    'creditlimit': ['mean', 'sum', 'count'],
    'name': 'count'
})

# Group statistics
print(df.groupby('industrycode')['creditlimit'].describe())

# Handle missing values